🔌
New: connect Claude & other AIs to GenAIList over MCP — research the catalog and contribute to the shared knowledge base. Learn how →
// RUN — FITS
Can the Instinct MI430X run VILA1.5-13B?
Yes — here's how fast and up to what context, with real measured numbers.
VILA1.5-13B on the Instinct MI430X
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 2252k | 543.9-664.8 t/s | ~0.9-1.1 (slow) |
| Q8 | yes | 2334k | 1087.8-1329.6 t/s | ~1.9-2.3 (slow) |
| Q4 | yes | 2371k | 1942.5-2374.2 t/s | ~3.3-4.1 (slow) |
Estimate (memory-bound). Beyond "max context" the KV cache spills to system RAM offload (depends on your system RAM) and speed drops to the "beyond" figure.
FAQ
Can the Instinct MI430X run VILA1.5-13B?▶
Yes. At Q4 it generates ~1942.5-2374.2 tok/s and fits up to 2371k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~3.3-4.1 tok/s.
How much context fits?▶
At Q4, up to about 2371k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.